Screenshots can’t be blocked
What happens
Anyone looking at your page can take a screenshot or record their screen.
Why
Once a picture is on someone’s screen, it belongs to their computer, not the website. No website, app or service can prevent this, whatever it claims.
What we do instead
Large views carry an invisible watermark unique to that viewer, so a screenshot can be traced back to the visit it came from. Your site also only ever shows display-sized images, so a screenshot is never your full-resolution original.
Invisible watermarks can be removed
What happens
Tracing doesn’t work on every copy.
Why
The watermark survives resizing, re-saving and small crops, but not:
- a whole-page screenshot (crop it to just the picture before tracing),
- a tiny thumbnail, or a copy that’s heavily edited, filtered or collaged,
- someone deliberately removing it. The watermarking technology is public, and so is a tool to remove it. Watermarks catch casual leaks, not determined ones.
Small gallery thumbnails aren’t watermarked at all, since they’re rarely worth leaking.
A technical person can get the real image from your site
What happens
Someone with programming skills can write a script that pretends to be a visitor on your site and reassembles the image.
Why
Any image your visitors can see, their browser can see, and so can a program that imitates a browser well enough. Domain locking makes this slow and fiddly and blocks it at scale; it isn’t a vault.
What we do
Known bots and scraping tools are refused, requests are rate-limited, and anything that doesn’t behave like a real visitor gets the scrambled copy.
The plain image option is weaker
What happens
If you use the plain image or image link (for places that don’t allow scripts), a visitor on your own page can right-click and save the real image.
Why
Without the script there is nothing to control what their browser does with the image.
What still works
Other sites, image search, bots and direct links still get the scrambled copy. Use the script embed wherever your site allows it.
Platforms that copy your image onto their own servers
What happens
Etsy, Amazon, Instagram, Facebook and most marketplaces and social networks make their own copy of any image you upload to them.
Why
After they copy it, the image is served from their servers, not ours, so we’re no longer involved.
What to do
Don’t post your original. In your dashboard, download a copy for social media made for that platform, and link to your own site for the full view. The copy is only as large as the platform shows and carries an invisible watermark for that platform, so a copy taken from your post can be traced to it.
What that still can’t do
Stop anyone saving or screenshotting your post, or tell you which person did. The watermark says which post a copy came from. It can be lost to heavy editing or removed on purpose, like any of our watermarks. Platforms also remove most of the copyright information stored inside a file.
We write the takedown notice; you send it
What happens
For a copy a patrol finds, we write the notice and point you to the right form. We don’t send it, and we can’t remove anything from someone else’s site.
Why
A takedown notice is a legal statement that you own the work and believe the use isn’t allowed, made under penalty of perjury. Only you can make it. The site decides what to do with it, and the person who posted the copy can dispute it.
What to do
Read the notice before sending it, and check the copy really is yours and isn’t a use you agreed to. Where the report says “possible copy” or “not confirmed”, look for yourself first. Our notice is general wording, not legal advice.
An upload record shows when, not who
What happens
An upload record is evidence that you held a file by a certain time. It doesn’t prove you created the picture.
Why
Anyone can upload a file they didn’t make. The record is one piece of evidence, alongside your camera originals, drafts and earlier posts. It also isn’t copyright registration, which in some countries gives you extra rights in court.
What to do
Keep the exact file you uploaded. The record matches that file only: a re-saved, resized or edited version has a different fingerprint. Images uploaded before records began have none; upload them again to get one.
A patrol only finds what search services have seen
A patrol, and the free image report, asks Google and TinEye where they’ve seen your picture. It can’t find:
- copies posted too recently for them to have crawled,
- copies behind a paywall or a sign-in: private groups, most social media posts, members-only sites, paid archives,
- copies changed heavily, such as large crops, collages and strong filters.
“No copies found” means none on record today, not none anywhere. To search, we send those services a reduced copy of the real image, and only after you allow it for that patrol.
The free report only sees your public pages
What happens
The free image report reads up to 12 pages of your site the way a visitor who isn’t signed in would. Pictures behind a paywall, a sign-in or a members’ area aren’t checked, nor are pictures drawn by scripts or set as page backgrounds.
Why
We can only search for what we can see, and we never sign in to your site.
What to do
Choose the report’s other option and upload up to 5 pictures with a free account, or sign up and patrol the images themselves. A patrol searches for the images you upload, wherever they’re shown.
A safety check only finds files it can reach
What happens
A safety check reads up to 100 pages of your site from its home page, its sitemap and, on WordPress, the public media library list. An original that nothing links to, a page behind a sign-in, or a picture drawn by a script or set as a page background isn’t checked. A heavily cropped or edited copy of your picture may not be recognised.
Why
We read your site the way a visitor who isn’t signed in would, and only once you’ve shown it’s yours with a DNS record.
What to do
After protecting your pictures, delete the originals from your media library or the folder on your server yourself. Use the check to catch what you missed, not as a list of everything that’s there.
Link previews on social media show the scrambled copy
What happens
When your page is shared on Facebook, X, LinkedIn or in a messaging app, the preview picture may appear scrambled.
Why
Those previews are fetched by bots, and bots only get the scrambled copy.
What to do
Set your page’s share image (og:image) to an unprotected picture, such as a
small or cropped version you’re happy to share.
Your images won’t appear in Google Images
What happens
Protected images are kept out of image search.
Why
That’s the point: image search is one of the main ways images are copied. But it also means people can’t find your work through image search.
What to do
Leave images you want found in search unprotected, and protect the ones that matter.
Visitors need JavaScript
What happens
A visitor with JavaScript turned off, or a very old browser, sees the scrambled copy.
Why
The real image is put together by a small script, and almost every browser runs scripts. The plain image option doesn’t need it.
Screen readers, layout and page speed
What happens
On your pages the real picture is drawn over the scrambled one by a small script, a moment after the page loads.
What still works
The image tag stays on your page, so its alt text is read by screen
readers as usual. Set the tag’s width and height and the page doesn’t jump when the picture
arrives. Images load as they scroll into view, at the size your page shows them.
What to expect
The first view of each image takes a fraction of a second longer, and a visitor who prints the page gets the scrambled copy.
Pages that hide where visitors come from
What happens
A page set to send no referrer information (Referrer-Policy: no-referrer)
shows the scrambled copy with the plain image option.
Why
The plain image checks which page it’s on using that information. The script embed isn’t affected.
We can unscramble your images
What happens
Our servers hold the key needed to show your real image on your sites.
Why
To show your real image to your visitors, the service has to be able to unscramble it. Your password protects the copies other people end up with, not your images from us. The key is stored encrypted, and we never store your original file.
If you run out of views, or we are down
What happens
Your protected images depend on our service each time a visitor views them. If our servers can’t be reached, visitors see the scrambled copy or an empty space. If your plan’s views are used up and the extra time we allow has passed, they see the scrambled copies until you have views again.
Why
The real image is never shown as a fallback, so nothing becomes unprotected.
What we do
When your views run out, your real images keep showing for up to 10 more days. We email you when that starts, 3 days before it ends and when it’s over, so you have time to move to a bigger plan or host those images somewhere else. The extra views are limited to what your plan averages over 10 days, a third of your monthly allowance, so a busy site can reach the limit sooner. They are counted and come out of your next monthly allowance. There’s no overage charge.
What to do
Watch the usage forecast in your dashboard, which shows when you’re likely to run out. Leave pictures that must always be visible, such as a logo or a hero image, unprotected. Service status shows whether we’re up.
Keep your own originals
What happens
A scrambled copy that someone has re-saved as JPEG or resized can no longer be unscrambled, even with your password.
Why
Unscrambling needs the exact pixels. That’s fine for protection, but Pixel Sentinel isn’t a backup service. Keep your original files. You can always download full-size originals from your dashboard.
Your scramble password can’t be reset by email
What happens
If you forget your scramble password, we can’t send you a reset link.
Why
It’s what keys your images, not your sign-in. You can still download your originals from the dashboard, but you’ll need the current password to change it or to unscramble a protected file. Keep it somewhere safe.
Changes take up to an hour to show everywhere
What happens
After you delete an image or change your banner, the old scrambled copy can still appear for up to an hour.
Why
Browsers and our delivery network keep copies of scrambled images for an hour to keep your pages fast.
Traffic numbers are close, not exact
What happens
Your traffic report may show fewer requests than you expect.
Why
Browsers keep images for a while, so a visitor who reloads a page doesn’t always ask us again, and “Save image as” usually saves the copy the browser already has without asking us. Sites and networks are what browsers report. Networks are shortened on purpose (the last part of the address is dropped), so they point to a household, office or mobile provider rather than one device.
Your visitors’ privacy
To make screenshots traceable and show traffic, we record for each view the page, a shortened network address, the country it’s in and the browser. Traffic records are kept for up to 180 days, and the records that let you trace a copy for two years after the view. We use them only for your traffic report and leak tracing, never for advertising or marketing, and we don’t sell them. Your own website’s privacy policy should mention this; the Data Processing Agreement has example wording.